Time to Revise ICH Q9

A change in terminology could emphasize patient protection.
Dec 02, 2012

The current International Conference on Harmonization (ICH) Q9, Quality Risk Management guideline was recommended for adoption by the three regulatory bodies of the United States, Japan, and the European Union on November 9, 2005 (1). The guideline's Expert Working Group (EWG) is to be commended for the content and quality of the current version.

The EWG's success is evidenced by the ready adoption by pharmaceutical companies, the industry in general, and by the many public presentations, training courses, and journal articles. Almost all pharmaceutical companies have a quality-risk management (QRM) program at some stage of development. The close relationship of ICH Q9 with ICH Q8 Pharmaceutical Development, and ICH Q10 Pharmaceutical Quality System has further solidified wide spread utilization of QRM in practical applications.

QRM and patient protection

Lynn D. Torbeck
Despite ICH Q9's success, it could benefit by revisions that place a stronger emphasis on minimizing risk to the patient rather than managing risk of quality. The current Q9 borrowed extensively from ISO/IEC Guide 73:2002 and has incorporated much of the format and concepts in ISO/IEC Guide 73:2002 (2). QRM, by implication, is internally focused, but there is a need to add an external focus to further minimize risk to patients.

ICH Q9 does comment on the role of the patient in the structure of the QRM program, but that context is limited. It is implied, but not actually said, that QRM means management of risk to the patient. The word patient occurs only seven times in the document; twice the patient is referred to as a stakeholder, twice regarding protection of the patient, once as an interested party, once regarding patient need, and once regarding high-quality drugs. Patient and risk are not mentioned together. By contrast, the phrase "Quality Risk Management" occurs 81 times.

It is assumed that the EWG intended that QRM would reduce risk to the patient, but the majority of industry readers may not recognize that assumption and may implement Q9 with an emphasis on management rather than on risk to the patient; obviously two different goals.

Q9 needs to be more explicit in its goal of patient risk minimization (PRM). This can be achieved by simply replacing the words "quality risk management" where they occur in the current version with "patient risk minimization." The shift from quality to patient and from management to minimization would refocus the entire quality program from internal to external and would further support the lifecycle concept of validation.

The risk to quality would still be managed by minimizing the risk to the patient. For example, a sentence in Q9 reads: "... the protection of the patient by managing the risk to quality should be considered of prime importance" (1).

Now consider the focus of this restatement: "The enhancement of quality by minimizing the risk to the patient should be considered the prime objective." Notice how this change shifts the focus from risk management to risk minimization. This revised statement has the additional benefit of implying continuous improvement and not just maintaining the status quo.

The shift from static management of quality to continuous reduction in patient risk, suggests the addition of statistically oriented tools and concepts. A new version would benefit greatly with these statistical inclusions:

  • definitions of probability for frequentists and Bayesian theory
  • a short review of how frequentists and Bayesian analysis differ
  • a note that in the field of statistics risk is often just a probability value
  • a short review of the decision risks as defined by Type I and Type II hypothesis errors
  • a short review of sampling risks as defined by acceptable quality limit (AQL) and by the limiting quality (LQ).

Note that in the field of statistics, the Type II error and the LQ are associated with the concept of patient or consumer risk. Minimizing these risks is inherent in good data design, collection, and analysis.


It is an opportune time to sharpen the focus on patient risk now that a measure of success with quality risk management has been achieved. Replacing the term "quality risk management" with "patient risk minimization" is one way to move to the next level of patient protection while building on the current strong base of ICH Q9.

Lynn D. Torbeck is a statistician at PharmStat Consulting, 2000 Dempster, Evanston, IL 60202, tel. 847.424.1314,
, http://www.PharmStat.com/.


1. ICH, Q9, Quality Risk Management (2005).

2. ISO/IEC Guide 73:2002.